On 20.06.2006, at 7:37 Uhr, Anjo Krank wrote:
So you need 200+(!) utterly boring components just for skinning. If you add functionality, you need to fix up 50 components.
Hmm. If you do it extensively, yes. But mostly you're done with different style sheets (the web designer's problem ... ;-)) and two or three different page wrappers. Not a real problem.
We have only a couple of things to change, so the components that vary are only the page wrapper (two different wrappers) and two other components. Out of about 100 components of this project. Not a real problem and the are much cleaner and easier to maintain then single "conditional monsters".
I really wouldn't do that for the estimated 50 customers - and if they have such different styles, that they really need 50 different html templates and can't go with say three or four and lots of style sheets, something is completely wrong there.
I don't want to track 200 mostly identical components and I don't want html work in the first place.
I don't want that too. And I don't do that!
So I'd do my very best to do everything inside of one components, by putting text fragments outside to the DB or files (the DB being able to quickly change things in deployment, the files have the benefit of easy/existing version control), do the gfx in css, do the html with conditionals or switch components and probably with a few javascripts for quick hacks.
This is very similar to what I do with the exception that I have two completely different page wrappers and a couple of pages with really different layout. Overall about 10 components I have to maintain. No real problem. The rest is done with css, images, some conditionals. Works great here.
cug _______________________________________________ Do not post admin requests to the list. They will be ignored. Webobjects-deploy mailing list (Webobjects-deploy@lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/webobjects-deploy/archive%40mail-archive.com This email sent to archive@mail-archive.com